More than ten million older Americans have diabetes and nearly all take medications to lower glucose levels. As a result of this treatment, many of these patients experience severe hypoglycemia, which is associated with poor outcomes including mortality. In order to improve the quality of diabetes care, we need to optimize treatment strategies and reduce these preventable complications. Yet, little information exists to help characterize a patient's risk for severe hypoglycemia. Clinicians and patients are essentially operating in the dark-without reliable evidence-when making important decisions about treatments with potentially life- threatening consequences. My goal is to generate that evidence and help older patients understand the risks they face so they can make more informed decisions about the trade-offs in treatment and, ultimately, achieve the outcomes that matter most to them. My hypothesis is that a simple clinical risk score can help identify which individuals are at greatest risk for severe hypoglycemia. Furthermore, this risk score can be incorporated into an electronic medical record system to improve treatment decisions and optimize diabetes management. My research has 3 specific aims: 1) estimate the annual incidence of severe hypoglycemia among older adults with type 2 diabetes (overall and by type of therapy); 2) develop and validate a clinical risk score for severe hypoglycemia; and 3) conduct a pilot study to assess the feasibility of using the risk score during clinical encounters and to assess its impact on treatment decisions. My career goals are to become an independent investigator and a national leader with expertise in outcomes research focusing on older adults who have diabetes. I am an endocrinologist with a track record of early success, including a GEMSSTAR R03 award. I have an exceptional team of mentors who are national leaders in outcomes research (Dr. Harlan Krumholz), geriatrics (Dr. Thomas Gill), and diabetes (Dr. Silvio Inzucchi). With their guidance, I have assembled an ideal set of data assets to address my aims. During the award period, I will gain skills in biostatistical methods, primary data collectio, and mixed methods research, and use implementation science methods to learn how to best bring my findings into clinical care. As part of my training, I will also lead the development of diabetes quality of care measures for the Centers of Medicare and Medicaid Services. The work proposed in my Beeson application is innovative because I intend to shift the paradigm of diabetes management among older individuals - from an intense focus on glucose control to a more balanced examination of patient-important outcomes, including adverse effects of therapies. I plan to forge strategic partnerships between leaders in aging research, outcomes research, and Optum Labs - which has an untapped database of extraordinary size and scope. This team can redefine the treatment of older patients with type 2 diabetes, by providing information that is currently nonexistent and yet critically needed.